Intelligent storage and supervision system for biological samples
By acquiring monitoring data sequences of the biological sample storage environment, calculating data anomaly, confidence, and external isolation, and combining anomaly feature confidence and enhancement coefficient, accurate early warning of the biological sample storage environment is achieved, solving the problem of false alarms in traditional systems and improving the accuracy of early warning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG KAIJING BIOTECHNOLOGY CO LTD
- Filing Date
- 2025-09-26
- Publication Date
- 2026-05-12
AI Technical Summary
Traditional biological sample storage monitoring systems, when using threshold methods for early warning, are prone to frequent false alarms due to instantaneous sensor errors and the opening of sample storage doors, making it difficult to accurately determine the actual abnormal situation in the environment.
The monitoring data sequence is acquired by the data acquisition module, and the data anomaly degree, confidence degree and external isolation degree are calculated by the first data analysis module. Combined with the anomaly feature confidence degree and enhancement coefficient of the second data analysis module, the environmental early warning module provides accurate early warning based on the anomaly score.
It improves the accuracy of environmental anomaly early warning, avoids frequent false alarms caused by sensor errors and warehouse door opening, and ensures timely warnings in real abnormal situations.
Smart Images
Figure CN121188404B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data monitoring technology, specifically to an intelligent storage and monitoring system for biological samples. Background Technology
[0002] In fields such as medical research, the storage and management of biological samples is particularly important. Vaccines, as a special type of biological sample, have their efficacy and safety closely related to storage conditions; environmental temperature and humidity have a significant impact on vaccine stability. Therefore, when storing vaccines and other biological samples, real-time monitoring of the storage environment is necessary to ensure the stability of the biological samples. Traditionally, monitoring the storage environment generally uses a threshold method, setting warning thresholds based on the storage requirements of the biological samples. When environmental parameters exceed the threshold, an alert is issued. While this method can quickly and easily alert to abnormal situations, it cannot determine whether the exceeding of the threshold represents a genuine anomaly, such as instantaneous sensor errors or temporary fluctuations in environmental parameters caused by opening the sample storage door. Such abnormal data does not reflect true environmental anomalies but leads to frequent alerts from the monitoring system, affecting managers' judgment of abnormal situations and even causing them to overlook real environmental anomalies. Therefore, relying solely on threshold-based monitoring and alerts is insufficient to accurately predict true environmental anomalies. Summary of the Invention
[0003] To address the aforementioned technical problems, the present invention aims to provide an intelligent storage and monitoring system for biological samples, the specific technical solution of which is as follows:
[0004] The data acquisition module is used to acquire monitoring data sequences from any sensor in the sample library environment;
[0005] The first data analysis module is used to obtain the data anomaly degree based on the data dispersion characteristics of the current moment in the monitoring data sequence; to obtain the data reliability based on the data difference characteristics between the current moment and adjacent moments, the recent data distribution characteristics of the current moment, and the data difference characteristics of adjacent sensors; and to obtain the external isolation degree based on the recent data change characteristics of the current moment, the difference change characteristics between the recent data of the current moment and the preset door opening time period data, and the time interval characteristics between the current moment and the door opening time.
[0006] The second data analysis module is used to obtain anomaly feature confidence based on the data credibility and the external isolation; obtain enhancement coefficient based on the changing trend characteristics of the monitored data sequence; and obtain anomaly score at the current moment based on the data anomaly, the anomaly feature confidence, and the enhancement coefficient.
[0007] An environmental early warning module is used to issue early warnings of environmental anomalies based on the anomaly score.
[0008] Furthermore, the step of obtaining the data anomaly degree based on the data discrete characteristics at the current moment in the monitored data sequence includes:
[0009] When the value at the current moment is not within the preset normal range, the absolute value of the difference between the value at the current moment and the nearest boundary value of the preset normal range is calculated to obtain the data anomaly degree at the current moment.
[0010] Furthermore, the step of obtaining data reliability based on the data difference characteristics between the current time and adjacent times, the recent data distribution characteristics of the current time, and the data difference characteristics of adjacent sensors includes:
[0011] In the monitoring data sequence, the absolute value of the difference between the current moment and the previous moment is calculated to obtain the instantaneous difference value; the standard deviation within a preset short window excluding the current moment is calculated to obtain a first value; the standard deviation within the preset short window including the current moment is calculated to obtain a second value; the ratio of the second value to the first value is calculated to obtain the fluctuation difference degree; the absolute value of the difference between the arbitrary sensor at the current moment and the other nearest sensor is calculated to obtain the local difference value; the product of the instantaneous difference value, the fluctuation difference degree, and the local difference value is calculated and negatively correlated to obtain the data reliability of the arbitrary sensor at the current moment.
[0012] Furthermore, the step of obtaining the external isolation degree based on the recent data change characteristics at the current moment, the difference change characteristics between the recent data at the current moment and the preset door opening time period data, and the time interval characteristics between the current moment and the door opening time includes:
[0013] The variation dispersion is obtained by summing the absolute values of the differences between the rate of change and the mean rate of change for each moment within the preset fluctuation period before the current moment; the standard deviation of the differences between all corresponding moments of the preset fluctuation period and the preset opening period data is calculated to obtain the difference variation degree; the external isolation degree is obtained by multiplying the time interval between the current moment and the nearest opening moment, the variation dispersion, and the difference variation degree.
[0014] Furthermore, the step of obtaining the confidence level of the abnormal features based on the data credibility and the external isolation includes:
[0015] The confidence level of the abnormal feature is obtained by multiplying the data confidence level by the external isolation level.
[0016] Furthermore, the step of obtaining the enhancement coefficient based on the changing trend characteristics of the monitoring data sequence includes:
[0017] The monitoring data sequence is linearly fitted, and the absolute value of the slope of the fitting result is used as the enhancement coefficient.
[0018] Further, the step of obtaining the anomaly score at the current moment based on the data anomaly degree, the anomaly feature confidence degree, and the enhancement coefficient includes:
[0019] In the formula, W represents the anomaly score. denoted by normalization, Q represents the data anomaly degree, K represents the enhancement coefficient, and F represents the confidence degree of the anomaly feature.
[0020] Furthermore, the step of issuing an environmental anomaly warning based on the anomaly score includes:
[0021] An alert is issued when the abnormal score exceeds a preset threshold, and the alert level is positively correlated with the abnormal score.
[0022] The present invention has the following beneficial effects:
[0023] In this invention, acquiring the data anomaly degree allows for a preliminary assessment of the anomaly level based on the current monitoring data. Since sensors are prone to momentary errors during monitoring, causing data to deviate significantly from the normal range and leading to false alarms of environmental anomalies, acquiring the data reliability degree characterizes the likelihood that the current data is caused by sensor momentary errors, further improving the accuracy of the alarm. Sample banks frequently experience door openings for biological sample retrieval. Door openings can cause environmental parameters in some areas to approximate the external environment, leading to false alarms of environmental anomalies. Therefore, acquiring the external isolation degree characterizes the extent to which the current data is affected by door openings, further improving the accuracy of the alarm. Acquiring the anomaly feature confidence degree accurately reflects the cause of the current data anomaly. Since the performance failure of environmental control equipment in the sample bank may exhibit slow-changing characteristics, acquiring the enhancement coefficient further corrects the anomaly feature confidence degree, improving the accuracy of the alarm. Finally, acquiring the anomaly score accurately determines the degree of environmental anomaly in the area where any sensor is located at the current moment. Environmental anomaly alarms based on the anomaly score improve the accuracy of the alarm and avoid frequent alarms caused by non-realistic environmental anomalies. Attached Figure Description
[0024] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0025] Figure 1 This is a block diagram of a smart storage and monitoring system for biological samples provided in one embodiment of the present invention. Detailed Implementation
[0026] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a biological sample intelligent storage and monitoring system proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.
[0028] The following description, in conjunction with the accompanying drawings, details the specific scheme of the intelligent storage and monitoring system for biological samples provided by this invention.
[0029] Please see Figure 1 The diagram illustrates a block diagram of a smart storage and monitoring system for biological samples according to an embodiment of the present invention. The system includes the following modules:
[0030] The data acquisition module S1 is used to acquire monitoring data sequences from any sensor in the sample library environment.
[0031] When biological samples are placed in a sample bank, appropriate temperature and humidity settings need to be set according to the characteristics of the biological samples for storage. Therefore, temperature and humidity sensors need to be installed in different areas of the sample bank to monitor the storage environment in real time and avoid abnormal temperature and humidity affecting the characteristics of the biological samples. In this embodiment of the invention, each temperature sensor obtains a corresponding temperature monitoring data sequence, and each humidity sensor obtains a corresponding humidity monitoring data sequence. The sequence length is the monitoring data within the most recent month at the current moment. The implementer can determine the collection duration and collection frequency according to the implementation scenario. In subsequent analysis, the monitoring data sequences of any sensor of any monitoring type, either temperature or humidity, are used for analysis.
[0032] The first data analysis module S2 is used to obtain the data anomaly degree based on the data dispersion characteristics of the current moment in the monitoring data sequence; to obtain the data reliability based on the data difference characteristics between the current moment and adjacent moments, the recent data distribution characteristics of the current moment, and the data difference characteristics of adjacent sensors; and to obtain the external isolation degree based on the recent data change characteristics of the current moment, the difference change characteristics between the recent data of the current moment and the preset door opening time period data, and the time interval characteristics between the current moment and the door opening time.
[0033] The more the storage environment parameters in the sample library deviate from the optimal storage parameters, the more likely an environmental anomaly is to occur at that moment. Therefore, the probability of anomaly at that moment is first determined, and the data anomaly degree is obtained based on the data dispersion characteristics of the current moment in the monitoring data sequence. Preferably, in this embodiment of the invention, the step of obtaining the data anomaly degree includes: if the value at the current moment is not within a preset normal range, it means that the storage environment may be abnormal; calculate the absolute value of the difference between the value at the current moment and the nearest boundary value of the preset normal range to obtain the data anomaly degree at the current moment; for example, if the temperature at the current moment is lower than the minimum value of the preset normal range, then calculate the absolute value of the difference between the temperature at the current moment and the minimum value; therefore, the larger the data anomaly degree, the more likely the storage environment is to be abnormal at the current moment; if the value at the current moment is within the preset normal range, then the data anomaly degree is 0, meaning that the storage environment is normal at the current moment, and no further calculation is performed. The implementer can determine the value of the preset normal range according to the implementation scenario.
[0034] Furthermore, significant data anomalies could be caused by malfunctions in environmental control equipment or other factors, such as instantaneous sensor errors or improper access to the sample storage facility. However, significant data anomalies not caused by environmental control equipment malfunctions do not affect the storage conditions of biological samples; therefore, warnings are unnecessary in such cases, thus avoiding frequent warnings and improving warning accuracy. When a sensor experiences instantaneous errors, the data at the current moment will fluctuate significantly, with an increased instantaneous rate of change, whereas under normal anomaly conditions, the data exhibits slow fluctuations. Instantaneous sensor errors also lead to significant differences compared to data from nearby sensors. Therefore, data reliability can be determined based on the differences between the current and adjacent time points, the recent data distribution characteristics at the current moment, and the differences between data from adjacent sensors.
[0035] Preferably, in this embodiment of the invention, the step of obtaining data reliability includes: calculating the absolute value of the difference between the current moment and the previous moment in the monitoring data sequence to obtain an instantaneous difference value; the larger the instantaneous difference value, the more obvious the local abrupt change, the less consistent it is with the characteristics of slow changes in environmental parameters, and the more likely it is caused by instantaneous sensor error. The standard deviation within a preset short window excluding the current moment is calculated to obtain a first value; in this embodiment of the invention, the preset short window is the monitoring data within the last 10 minutes. The standard deviation within the preset short window including the current moment is calculated to obtain a second value; if the sensor has an instantaneous error at the current moment, it will cause significant data fluctuation characteristics within the preset short window, increasing the standard deviation. The ratio of the second value to the first value is calculated to obtain the fluctuation difference degree; when the second value is larger than the first value, it means that the value at the current moment will cause increased data fluctuation characteristics within the preset short window; the larger the fluctuation difference degree, the more likely the sensor has an instantaneous error at the current moment. Calculate the absolute value of the difference between the current sensor and the nearest other sensor to obtain the local difference value. A larger local difference value indicates a greater data difference between adjacent sensors, making instantaneous sensor errors more likely. Since the opening of the sample library can also cause differences in data from adjacent sensors, simultaneously analyzing instantaneous difference, fluctuation difference, and local difference values can accurately determine the probability of instantaneous errors in the current sensor. Calculate the product of instantaneous difference, fluctuation difference, and local difference values and apply a negative correlation mapping to obtain the data reliability of the current sensor. Smaller instantaneous difference, fluctuation difference, and local difference values indicate more accurate and reliable data from the sensor, making it less likely that instantaneous errors are causing significant data anomalies. Formulas for obtaining data reliability include:
[0036]
[0037] In the formula, R represents the data reliability, and H represents the instantaneous difference value. Indicates the second value. Indicates the first value. G represents the degree of fluctuation difference, and G represents the local difference value. This represents an exponential function with the natural constant as its base.
[0038] Furthermore, due to the significant differences between the environment inside the sample bank and the external environment, environmental parameters in some areas will change after the door is opened, leading to large data discrepancies. Moreover, the act of opening the door to access biological samples is quite common, easily triggering unnecessary warnings. When the door is opened, the storage environment parameters change slowly, gradually approaching the external environment. Secondly, the opening of the sample bank is a long-term occurrence, and the changes in storage environment parameters after each opening are relatively similar. Furthermore, the closer to the opening time, the greater the likelihood of being affected by the opening. Therefore, the degree of external isolation can be obtained based on the recent data change characteristics at the current moment, the difference in recent data between the current moment and the data during the preset opening period, and the time interval between the current moment and the opening moment.
[0039] Preferably, in this embodiment of the invention, the step of obtaining the external isolation degree includes: calculating the sum of the absolute values of the differences between the rate of change and the mean rate of change at each moment within a preset fluctuation period prior to the current moment, to obtain the variation dispersion; in this embodiment of the invention, the rate of change is the difference between the values at any moment and the previous moment divided by the time interval between the two; if the rate of change at the current moment is positive, the trough between the current moment and the most recent peak is taken as the starting point of the preset fluctuation period, and the current moment is taken as the ending point; if the rate of change at the current moment is negative, the trough between the two most recent peaks at the current moment is taken as the starting point of the preset fluctuation period, and the current moment is taken as the ending point. The smaller the difference between the rate of change at each moment and the mean rate of change of the preset fluctuation period, the smoother the change, which is more consistent with the changes in storage environment parameters caused by door opening behavior, and less consistent with the situation of abnormal fluctuations in environmental parameters; therefore, the smaller the variation dispersion, the more likely the current moment is to be affected by door opening behavior; the larger the variation dispersion, the more likely the environmental control equipment is malfunctioning. The standard deviation of the difference between the data of the preset fluctuation period and the preset door opening period is calculated to obtain the degree of difference variability. The preset door opening period data is collected before the door opening period. The preset fluctuation period data and the preset door opening period data are aligned from the starting position, and the standard deviation of the difference between each corresponding moment is calculated. Due to different external environments, the data of the preset fluctuation period and the preset door opening period data may not be exactly the same, but the data change patterns are relatively similar. Therefore, the smaller the degree of difference variability, the more similar the data differences at each moment, and the more likely the preset fluctuation period is caused by door opening behavior. The product of the time interval between the current moment and the most recent door opening moment, the degree of variation dispersion, and the degree of difference variability is calculated to obtain the external isolation degree. If the time interval between the current moment and the most recent door opening moment is smaller, it means that the current moment is more likely to be affected by door opening behavior. Therefore, the smaller the external isolation degree, the more likely the current moment is affected by the external environment of door opening behavior. The larger the external isolation degree, the less affected by door opening behavior, and the more likely it is caused by a real abnormality of the environmental control equipment.
[0040] The second data analysis module S3 is used to obtain the confidence level of abnormal features based on data credibility and external isolation; to obtain the enhancement coefficient based on the changing trend characteristics of the monitored data sequence; and to obtain the anomaly score at the current moment based on the data anomaly degree, the confidence level of abnormal features, and the enhancement coefficient.
[0041] After obtaining the data reliability and external isolation of any sensor at the current moment, the confidence level of the abnormal feature can be obtained based on the data reliability and external isolation. Preferably, in this embodiment of the invention, the step of obtaining the confidence level of the abnormal feature includes: calculating the product of the data reliability and the external isolation to obtain the confidence level of the abnormal feature. The higher the data reliability, the lower the possibility of the sensor's instantaneous error affecting the signal. The higher the external isolation, the lower the possibility of the door opening behavior affecting the signal. Therefore, the higher the confidence level of the abnormal feature, the higher the possibility of a real abnormality in the storage environment, and the greater the need for early warning.
[0042] Furthermore, since performance failures of environmental control equipment may occur slowly, such as a gradual weakening of cooling capacity or humidity enhancement capacity, these performance failures are difficult to detect in the early stages, but in the long term, the monitoring data shows a slight changing trend; and the probability of the failure severity increasing gradually increases in the later stages. Therefore, this characteristic can be used to further correct the confidence level of the abnormal features. Thus, the enhancement coefficient can be obtained based on the changing trend characteristics of the monitoring data sequence. Preferably, in this embodiment of the invention, the step of obtaining the enhancement coefficient includes: performing linear fitting on the monitoring data sequence, and using the absolute value of the slope of the fitting result as the enhancement coefficient. Under normal circumstances, the data after linear fitting of the monitoring data sequence is relatively stable and will not show a gradual increasing or decreasing trend, at which point the slope tends to 0. If the slope is not 0, it means that there is a slight changing trend in the monitoring data sequence, and the probability of performance failure of the environmental control equipment increases. When the absolute value of the slope is larger, the enhancement coefficient is larger, the trend of equipment failure is more obvious, and the possibility of a real anomaly at the current moment is greater. Then, the anomaly score at the current moment can be obtained based on the data anomaly degree, the confidence level of the abnormal features, and the enhancement coefficient. Preferably, in this embodiment of the invention, the step of obtaining the anomaly score includes:
[0043]
[0044] In the formula, W represents the anomaly score. Let represent normalization, Q represent the data anomaly degree, K represent the enhancement coefficient, and F represent the confidence level of the anomaly feature. Since the changes in the monitoring data sequence caused by the slow changes in performance faults are relatively weak, the enhancement coefficient is small and may even be 0. Therefore, [the following is a more accurate translation:] ... The confidence level of anomaly features is enhanced. A higher anomaly level, enhancement coefficient, and anomaly feature confidence level at the current moment indicate a greater likelihood of real-world environmental anomalies in the area where any sensor is located. A higher anomaly score necessitates a stronger warning. Conversely, even if the current data anomaly level is high, a lower enhancement coefficient and anomaly feature confidence level result in a lower anomaly score, suggesting a lower probability of real-world environmental anomalies and thus avoiding frequent warnings.
[0045] The environmental early warning module S4 is used to provide early warnings of environmental anomalies based on anomaly scores.
[0046] After obtaining the anomaly score at the current moment, environmental anomaly warnings can be issued based on the anomaly score. Specifically, a warning is issued when the anomaly score exceeds a preset threshold, and the warning level is positively correlated with the anomaly score. In this embodiment of the invention, the warning level is divided into four levels. The higher the anomaly score, the higher the level, and the more severe the anomaly. The response measures change with the warning level. Implementers can determine the preset threshold and warning level according to the implementation scenario, which is not limited here. Managers can inspect and implement corresponding response measures for the environmental control equipment in the area where the sensor that issued the warning is located based on the warning situation. Using anomaly scores for warnings can avoid frequent warnings caused by instantaneous sensor errors and common warehouse door opening behaviors, improve the accuracy of warnings, and thus achieve accurate warnings in real anomaly situations.
[0047] In summary, this invention provides an intelligent storage and monitoring system for biological samples. It obtains data anomaly degree based on the discrete characteristics of data at the current moment in the monitoring data sequence; data reliability based on the differences between the current moment and adjacent moments, recent data distribution characteristics, and differences between data from adjacent sensors; external isolation based on recent data change characteristics, differences between recent data and data from a preset door opening period, and the time interval between the current moment and the door opening moment; anomaly feature confidence level based on data reliability and external isolation level; and enhancement coefficient based on the changing trend characteristics of the monitoring data sequence. This invention obtains anomaly scores for the current moment based on data anomaly degree, anomaly feature confidence level, and enhancement coefficient, and performs environmental anomaly early warning, thus improving the accuracy of environmental anomaly early warning.
[0048] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0049] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.
Claims
1. A biological sample intelligent storage and monitoring system, characterized in that, The system includes the following modules: The data acquisition module is used to acquire monitoring data sequences from any sensor in the sample library environment; The first data analysis module is used to obtain the data anomaly degree based on the data dispersion characteristics of the current moment in the monitoring data sequence; to obtain the data reliability based on the data difference characteristics between the current moment and adjacent moments, the recent data distribution characteristics of the current moment, and the data difference characteristics of adjacent sensors; and to obtain the external isolation degree based on the recent data change characteristics of the current moment, the difference change characteristics between the recent data of the current moment and the preset door opening time period data, and the time interval characteristics between the current moment and the door opening time. The second data analysis module is used to obtain anomaly feature confidence based on the data credibility and the external isolation; obtain enhancement coefficient based on the changing trend characteristics of the monitored data sequence; and obtain anomaly score at the current moment based on the data anomaly, the anomaly feature confidence, and the enhancement coefficient. The environmental early warning module is used to provide early warnings of environmental anomalies based on the anomaly score. The step of obtaining the external isolation degree based on the recent data change characteristics at the current moment, the difference change characteristics between the recent data at the current moment and the preset door opening time period data, and the time interval characteristics between the current moment and the door opening time includes: The variation dispersion is obtained by summing the absolute values of the differences between the rate of change and the mean rate of change at each moment within the preset fluctuation period before the current moment. Calculate the standard deviation of the difference between all corresponding moments of the preset fluctuation period and the preset opening period data to obtain the degree of difference variation; calculate the product of the time interval between the current moment and the nearest opening moment, the degree of variation, and the degree of difference variation to obtain the degree of external isolation.
2. The intelligent storage and monitoring system for biological samples according to claim 1, characterized in that, The step of obtaining the data anomaly degree based on the data discrete characteristics at the current moment in the monitoring data sequence includes: When the value at the current moment is not within the preset normal range, the absolute value of the difference between the value at the current moment and the nearest boundary value of the preset normal range is calculated to obtain the data anomaly degree at the current moment.
3. The intelligent storage and monitoring system for biological samples according to claim 1, characterized in that, The step of obtaining data reliability based on the data difference characteristics between the current time and adjacent times, the recent data distribution characteristics of the current time, and the data difference characteristics of adjacent sensors includes: In the monitoring data sequence, the absolute value of the difference between the current moment and the previous moment is calculated to obtain the instantaneous difference value; the standard deviation within a preset short window excluding the current moment is calculated to obtain a first value; the standard deviation within the preset short window including the current moment is calculated to obtain a second value; the ratio of the second value to the first value is calculated to obtain the fluctuation difference degree; the absolute value of the difference between the arbitrary sensor at the current moment and the other nearest sensor is calculated to obtain the local difference value; the product of the instantaneous difference value, the fluctuation difference degree, and the local difference value is calculated and negatively correlated to obtain the data reliability of the arbitrary sensor at the current moment.
4. The intelligent storage and monitoring system for biological samples according to claim 1, characterized in that, The step of obtaining the confidence level of the abnormal features based on the data credibility and the external isolation includes: The confidence level of the abnormal feature is obtained by multiplying the data confidence level by the external isolation level.
5. The intelligent storage and monitoring system for biological samples according to claim 1, characterized in that, The step of obtaining the enhancement coefficient based on the changing trend characteristics of the monitoring data sequence includes: The monitoring data sequence is linearly fitted, and the absolute value of the slope of the fitting result is used as the enhancement coefficient.
6. The intelligent storage and monitoring system for biological samples according to claim 1, characterized in that, The step of obtaining the anomaly score at the current moment based on the data anomaly degree, the anomaly feature confidence degree, and the enhancement coefficient includes: In the formula, W represents the anomaly score. denoted by normalization, Q represents the data anomaly degree, K represents the enhancement coefficient, and F represents the confidence degree of the anomaly feature.
7. The intelligent storage and monitoring system for biological samples according to claim 1, characterized in that, The step of providing environmental anomaly early warning based on the anomaly score includes: An alert is issued when the abnormal score exceeds a preset threshold, and the alert level is positively correlated with the abnormal score.